Abstract We present results from an observational campaign on the close binary system 2MASS J16211735+4412541 and a preliminary model based on the photometric data gathered during the quiescent and outburst levels. The modeling, done with the Wilson–Devinney code and its improvements, failed to reproduce the observational properties of the system. A secondary minimum obtained within the stellar model that is too shallow, as well as the evidence provided by the spectroscopic observations performed at outburst and quiescence, point toward an accretion disk surrounding one component, likely a white dwarf, as the cause of the outburst. Using a simple disk model, we modeled the observed multicolor light curves taken two (2016 August) and eight (2017 March) months after the outburst. We obtained a reasonable fit to the 2016 August light curves but those from 2017 March cannot be explained with the same parameters. We conclude that J1621 is an eclipsing cataclysmic binary, with an accretion disk still present almost a year after outburst, and not a contact-type system as previously classified. The binary is seen at an inclination of about 84° and there is evidence of changing accretion rates and disk parameters as a result of the outburst. Our results indicate that more cataclysmic variables may be hidden among contact binaries.